[mpi3-coll] [Mpi-forum] MPI_Comm_create_group Meeting

Fri Dec 9 10:30:13 CST 2011

Thanks for the feedback.  I've added the text for ticket #305 to the 
ticket.  I'll also update the schedule to include a formal reading of 
ticket #305 in conjunction with ticket #286; I think we should be able 
to squeeze it in the same slot to save time.

Cheers,
  ~Jim.

On 12/8/11 7:32 PM, Bronis R. de Supinski wrote:
>
> Jim:
>
> The new ticket looks fine to me. I suggest we move it
> forward so we can have a reading in January (maybe some
> others can review it before the telecon). The updated
> PDF also looks fine. I was wondering if we have strong
> enough language on ordering requirements but I have
> convinced myself that the words with the tags discussion
> are sufficient.
>
> Bronis
>
>
>
> On Thu, 8 Dec 2011, Jim Dinan wrote:
>
>> Hi Bronis,
>>
>> Thanks very much for the feedback. I've created a ticket (#305) with
>> some initial text for MPI_Intercomm_create:
>>
>> https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/305
>>
>> I updated the MPI_Comm_create_group text with your corrections and also
>> made some additional updates to the tags paragraph (page 228, line 12)
>> to hopefully address a few other questions I got off-list. A new PDF is
>> attached. I'd appreciate a second look if you have time.
>>
>> Thanks,
>> ~Jim.
>>
>> On 12/7/11 9:00 PM, Bronis R. de Supinski wrote:
>>>
>>> Jim:
>>>
>>> I think it would be good to create a companion ticket to make
>>> MPI_Intercomm_create use the tag space. We should get that
>>> ticket ready now (I expect it is just a few lines of change)
>>> so that it will be in MPI 3.0, assuming ticket 286 passes.
>>>
>>> Anyway, please change:
>>>
>>> "new intracommunicator, newcomm, whose communication group is deﬁned by
>>> the group argument."
>>>
>>> to:
>>>
>>> "new intracommunicator, newcomm, for which the group argument defines
>>> the communication group."
>>>
>>> Also, please change:
>>>
>>> "to order calls and/or use diﬀerent tag arguments"
>>>
>>> to
>>>
>>> "to order calls and/or to use diﬀerenttag arguments"
>>>
>>> Bronis
>>>
>>>
>>>
>>> On Wed, 7 Dec 2011, Jim Dinan wrote:
>>>
>>>> Hi All,
>>>>
>>>> I've updated the text with feedback from the last meeting, adding text
>>>> to define a collective tag space. Please take a look at the ticket286
>>>> edits in the attached PDF and send any feedback.
>>>>
>>>> Also, I'd like to have another conference call early next week to
>>>> discuss and get ready for a reading in January. Please fill in your
>>>> availability on the doodle poll:
>>>>
>>>> http://www.doodle.com/yx5b2f5f5hwc2qzq
>>>>
>>>> Thanks,
>>>> ~Jim.
>>>>
>>>> On 11/8/11 3:57 PM, James Dinan wrote:
>>>>> Hi All,
>>>>>
>>>>> Notes from yesterday's meeting are attached. Please note that we are
>>>>> moving all future discussion of this ticket to the mpi3-coll mailing
>>>>> list.
>>>>>
>>>>> Best,
>>>>> ~Jim.
>>>>>
>>>>> In attendance:
>>>>>
>>>>> Adam Moody, David Goodell, Jim Dinan, Pavan Balaji, Fab Tiller, Bronis
>>>>> de Supinski, Jeff Hammond
>>>>>
>>>>> Item 1: Name of the routine.
>>>>>
>>>>> We decided to stick with MPI_Comm_create group as the best way of
>>>>> matching the MPI_CLASS_ACTION_SUBSET naming convention.
>>>>>
>>>>> Item 2: Tag versus color.
>>>>>
>>>>> We decided to use a tag, which is defined to be in a collective tag
>>>>> space that is distinct from the point-to-point tag space. The
>>>>> collective
>>>>> tag will follow all MPI tag semantics, however it will not conflict
>>>>> with
>>>>> any point-to-point tags (including MPI_ANY_TAG). This essentially
>>>>> exposes the expected implementation in the standard, providing more
>>>>> flexible semantics to the user. In addition, it allows an
>>>>> implementation
>>>>> to make more efficient use of the tag space by defining tag conflicts
>>>>> across tagged collective operations (currently only
>>>>> MPI_Intercomm_create
>>>>> and MPI_Comm_create_group, but we could define others in the future).
>>>>>
>>>>> In a separate ticket, we can update MPI_Intercomm_create to specify
>>>>> that
>>>>> its tag is also in the collective tag space. This will be backward
>>>>> compatible, but will make MPI_Intercomm_create considerably easier to
>>>>> use because its tags will no longer conflict with point-to-point
>>>>> operations.
>>>>>
>>>>> Tasks:
>>>>>
>>>>> * Jim will work on a revision of the text that follows the above
>>>>> semantics.
>>>>>
>>>>> * The group will reconvene after Thanksgiving to review new text.
>>>>>
>>>>> * The group will move discussion to the collectives mailing list.
>>>>>
>>>>>
>>>>> On 11/4/11 11:07 AM, James Dinan wrote:
>>>>>> Based on the doodle poll, it looks like Monday at 3pm CT will be the
>>>>>> best time to meet. We'll use the hybrid WG telecon bridge:
>>>>>>
>>>>>> Domestic dial number: 866-654-6744
>>>>>> International dial number: 517-308-8226
>>>>>> Passcode: 6086006
>>>>>>
>>>>>> See you then,
>>>>>> ~Jim.
>>>>>>
>>>>>> On 11/1/11 2:33 PM, James Dinan wrote:
>>>>>>> Hi All,
>>>>>>>
>>>>>>> I would like to form a group to review the MPI_Comm_create_group()
>>>>>>> proposal (#286) for a formal reading at the next Forum meeting.
>>>>>>> If you
>>>>>>> are interested in joining the discussion, could you please fill in
>>>>>>> your
>>>>>>> availability for a one hour conference call next week:
>>>>>>>
>>>>>>> http://www.doodle.com/9tganzq2hqiqy7kt
>>>>>>>
>>>>>>> We have a couple issues to discuss:
>>>>>>>
>>>>>>> 1.) The name of the function: MPI_Comm_create_group,
>>>>>>> MPI_Group_comm_create, MPI_Comm_create_from_group, ...
>>>>>>>
>>>>>>> 2.) The tag argument: Should we use a tag or a "color" to
>>>>>>> establish a
>>>>>>> safe communication conduit on the parent communicator. As
>>>>>>> discussed at
>>>>>>> the Forum, using a tag could conflict with the use of MPI_ANY_TAG
>>>>>>> on the
>>>>>>> parent communicator. A color would avoid this issue, however it
>>>>>>> could
>>>>>>> complicate tag matching.
>>>>>>>
>>>>>>> 3.) Changes to the text.
>>>>>>>
>>>>>>> Thanks,
>>>>>>> ~Jim.
>>>>>>>
>>>>>>> PS- I've attached a copy of the current text with comments from the
>>>>>>> formal reading at the last Forum meeting (in Acrobat: Comments ->
>>>>>>> Comments View -> Show Comments List for the full list).
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> mpi-forum mailing list
>>>>>>> mpi-forum at lists.mpi-forum.org
>>>>>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
>>>>>> _______________________________________________
>>>>>> mpi-forum mailing list
>>>>>> mpi-forum at lists.mpi-forum.org
>>>>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
>>>>> _______________________________________________
>>>>> mpi-forum mailing list
>>>>> mpi-forum at lists.mpi-forum.org
>>>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
>>>>
>>>
>>>
>>> _______________________________________________
>>> mpi3-coll mailing list
>>> mpi3-coll at lists.mpi-forum.org
>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-coll
>>
>
>
> _______________________________________________
> mpi3-coll mailing list
> mpi3-coll at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-coll